fotros mohammadhasan; mostafa omidali; amirmohammad galavani
Abstract
The aim of this study is to estimate the domestic balance of natural gas per capita in the Iran, as well as its forecast for the period 2017 - 2037. In this study, with employing dynamic models Autoregressive Distributed Lag (ARDL), at first, long-term and short-term elasticity of per capita natural ...
Read More
The aim of this study is to estimate the domestic balance of natural gas per capita in the Iran, as well as its forecast for the period 2017 - 2037. In this study, with employing dynamic models Autoregressive Distributed Lag (ARDL), at first, long-term and short-term elasticity of per capita natural gas demand in Iran for the period 1981-2016 is estimated. Then with using a hybrid ARDL and ARIMA model, we predict the balance natural gas per capita up to the year 2037. The results show that amount of per capita natural gas demand will reach 4177.36 million cubic meters in 2037, as well as the amount of per capita natural gas supply will reach 3417.26 million cubic meters in this years. For responding this excess demand should be adopting policies to increase production or constrainting natural gas demand.
Mansour Zaraanjad; pouyan kiani; Salah Ebrahimi; Ali Raoofi
Volume 2, Issue 5 , January 2013, , Pages 107-207
Abstract
Crude oil prices are influenced by many factors. Inclusion of all these determinants in a single model is complex and inefficient. In this case, using time series approach might be appropriate. In the later method past behavior of oil prices is used to forecast its future volatility. Several time series ...
Read More
Crude oil prices are influenced by many factors. Inclusion of all these determinants in a single model is complex and inefficient. In this case, using time series approach might be appropriate. In the later method past behavior of oil prices is used to forecast its future volatility. Several time series studies were conducted to forecast oil prices using methods such as autoregressive integrated moving average (ARIMA) models and artificial neural networks (ANN). All these methods need a large volume of data to have accurate forecasting. One way to overcome this limitation is to use fuzzy regression (FA) models which can give more accurate forecasting with less data. In this study, the three methods, fuzzy regression, ARIMA and fuzzy autoregressive integrated moving average (FARIMA) were applied using the daily oil price in order to forecast oil prices. To compare the forecast accuracy of the model, the prediction error criteria was used. The results showed that the performance of FARIMA is much better than the other two models.